Human milk oligosaccharides (HMOs) have been the subject of much interest in the past few years. In particular, commercialization efforts for the synthesis of these complex carbohydrates including secreted oligosaccharides have increased significantly due to their roles in numerous biological processes occurring in the human organism. One prominent natural human source of such complex oligosaccharides is mammalian milk. Mammalian milk contains up to 10% carbohydrate, of which the disaccharide, lactose (Gal(β1-4)Glc), is usually a prominent component. Milk and colostrum also contain lesser amounts of other saccharides, referred to as milk oligosaccharides, nearly all of which have a lactose unit at their reducing end to which GlcNAc, Gal, Fuc and/or Neu5Ac or Neu5Gc residues can be attached (Messer and Urashima, 2002, Trends Glycosci. Glycotech, 14, 153-176; and Urashima et al., Advanced Dairy Chemistry, Volume 3: Lactose, Water, Salts and Minor Constituents, 2009, pp. 295-349).
To date, the structures of at least 115 oligosaccharides of human milk have been determined, while mass spectra (MS) data have suggested the presence of almost 130 oligosaccharides in human milk or colostrums (Newburg and Neubauer, 1995, Carbohydrates in milks: Analysis, quantities and significance. In: Handbook of Milk Composition (R. G. Jensen, ed.), pp. 273-249, Academic Press, San Diego, USA). Moreover, matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOFMS) analyses suggest that polysaccharides, consisting of more than 50 monosaccharide residues, as indicated by size exclusion chromatography, are also present in human milk. Therefore, considerably more than 130 different saccharides are probably present in human milk (see also Urashima et al., Advanced Dairy Chemistry, Volume 3: Lactose, Water, Salts and Minor Constituents, 2009, pp. 295-349; and TADASU URASHIMA et al, MILK OLIGOSACCHARIDES, Nova Biomedical Books, New York, 2011, ISBN: 978-1-61122-831-1).
The 115 human milk oligosaccharides, the structures of which have been determined to date, can be grouped into 13 series based on their core structures. Such 13 core structures are exemplarily shown in Table 1 below:
TABLE 113 different core structures of human milk oligosaccharides (HMOs)NoCore nameCore structure1lactose (Lac)Galβ1-4Glc2lacto-N-tetraose (LNT)Galβ1-3GlcNAcβ1-3Galβ1-4Glc3lacto-N-neotetraoseGalβ1-4GlcNAcβ1-3Galβ1-4Glc(LNnT)4lacto-N-hexaose (LNH)Galβ1-3GlcNAcβ1-3(Galβ1-4GlcNAcβ1-6)Galβ1-4Glc5lacto-N-neohexaoseGalβ1-4GlcNAcβ1-3(Galβ1-4GlcNAcβ1-6)Galβ1-4Glc(LNnH)6para-lacto-N-hexaoseGalβ1-3GlcNAcβ1-3Galβ1-4GlcNAcβ1-3Galβ1-4Glc(para-LNH)7para-lacto-N-Galβ1-4GlcNAcβ1-3Galβ1-4GlcNAcβ1-3Galβ1-4Glcneohexaose (para-LNnH)8lacto-N-octaose (LNO)Galβ1-3GlcNAcβ1-3(Galβ1-4GlcNAcβ1-3Galβ1-4GlcNAcβ1-6)Galβ1-4Glc9lacto-N-neooctaoseGalβ1-4GlcNAcβ1-3(Galβ1-3GlcNAcβ1-3Galβ1-4GlcNAcβ1-(LNnO)6)Galβ1-4Glc10Iso-lacto-N-octaoseGalβ1-3GlcNAcβ1-3(Galβ1-3GlcNAcβ1-3Galβ1-4GlcNAcβ1-(iso-LNO)6)Galβ1-4Glc11para-lacto-N-octaoseGalβ1-3GlcNAcβ1-3Galβ1-4GlcNAcβ1-3Galβ1-4GlcNAcβ1-(para-LNO)3Galβ1-4Glc12lacto-N-neodecaoseGalβ1-3GlcNAcβ1-3[Galβ1-4GlcNAcβ1-3(Galβ1-4GlcNAcβ1-(LNnD)6)Galβ1-4GlcNAcβ1-6]Galβ1-4Glc13lacto-N-decaose (LND)Galβ1-3GlcNAcβ1-3[Galβ1-3GlcNAcβ1-3(Galβ1-4GlcNAcβ1-6)Galβ1-4GlcNAcβ1-6]Galβ1-4Glc
As found by Urashima et al. (see also Urashima et al., Advanced Dairy Chemistry, Volume 3: Lactose, Water, Salts and Minor Constituents, 2009, pp. 295-349; and TADASU URASHIMA et al, MILK OLIGOSACCHARIDES, Nova Biomedical Books, New York, 2011, ISBN: 978-1-61122-831-1) the many variations of the oligosaccharides are constructed by the addition of a Neu5Acβ2-3/2-6 residue to Gal or GlcNAc, and of Fucα1-2/1-3/1-4 to Gal, GlcNAc or a reducing Glc of the core units. The main structural features of human milk oligosaccharides are the presence of oligosaccharides containing the type I unit (Gal(β1-3)GlcNAc), as well as those containing the type II unit (Gal(β1-4)GlcNAc), and oligosaccharides containing the type I predominate over those containing the type II unit. The milk oligosaccharides of other species investigated to date mostly exhibit the type II but not the type I unit.
The large variety of oligosaccharides in human milk and colostrum and the difference to other species, however, makes it difficult to prepare suitable replacements in foods, particularly in infant food formulae, which display at least part of the entire spectrum of human milk oligosaccharides. Furthermore, their recognized importance in the maturation of the immune system and their prognostic use as immunomodulators underlines their importance as a possible immunomodulator.
Accordingly, there is an urgent need in the art for the preparation of complex oligosaccharides and mixtures thereof, which resemble as much as possible or even reproduce the variety of complex oligosaccharides in human milk.
Many attempts have been carried out in this respect to produce individual HMOs via organo-chemical synthesis and, due to its stereoselectivity, via enzymatic means. Enzymatic means have been increasingly explored in the last two decades.
Notably, in biological systems, Leloir-type glycosyltransferases (GTs, EC 2.4.1.-) and glycosidases (also called glycoside hydrolases: GHs, EC 3.2.1.-) constitute the two major classes of carbohydrate-processing enzymes, which may be utilized in the production of HMOs. Both classes of enzymes act to transfer a glycosyl group from a donor to an acceptor resulting in oligosaccharide production. The use of glycosyltransferases for synthesis in industrial processes, however, is limited both by the availability of the desired enzymes due to problems with expression and solubility and the high costs of the activated donor sugars. These nucleotide donors may be typically generated in situ, but the process requires additional enzymes (see Hanson, S., et al., Trends Biochem Sci, 2004. 29(12): p. 656-63). In contrast to glycosyltransferases, glycosidases have a wide range of donor substrates employing usually monosaccharides, oligosaccharides or/and engineered substrates (i.e. substrates carrying various functional groups). They often display activity towards a large variety of carbohydrate and non-carbohydrate acceptors. Another advantage of the use of glycosidases compared to glycosyltransferases is their robustness and accessibility.
In vivo, glycosidases usually catalyze the hydrolysis of glycosidic linkages with either retention or inversion of stereochemical configuration in the product. In vitro, they can catalyse the formation of a new glycosidic bond either by transglycosylation or by reverse hydrolysis (ie. condensation). Under kinetically controlled reactions these enzymes (typically, retaining glycosidases) can be used to form glycosidic linkages using a glycosyl donor activated by a good anomeric leaving group (e.g. nitrophenyl glycoside). In contrast, the thermodynamically controlled reverse hydrolysis uses high concentrations of free sugars. However, even though the appropriate application of glycosidases in the synthetic direction is of considerable interest, it remains challenging as optimal conditions and suitable substrates have to be found to drive the reaction in the desired direction and to avoid hydrolysis of the products.
Another approach to overcome this bottleneck and to make glycosidases more suitable for oligosaccharide synthesis has been recently developed by providing modified enzymes (variants). Thus, during these two past decades, protein engineering based on rational or combinatorial techniques has proven to be extremely powerful to generate biocatalysts with improved transglycosylation activity and efficiency.
However, even though many organo-chemical syntheses or enzyme based syntheses for basic human milk oligosaccharide structures or their precursors have been published meanwhile (e.g. for the synthesis of some individual sialylated HMOs or HMO benzyl/substituted benzyl glycosides using a trans-sialidase and 3′-SL see WO 93/18787 and WO 2012/007588), such synthesis methods still do not allow the preparation of complex mixtures of naturally occurring oligosaccharides or derivatives thereof. Preparing such mixtures on the basis of individually designed syntheses of single HMOs is furthermore costly and may not resemble the large variety of naturally occurring HMOs.
Accordingly, it is an object underlying the present invention to provide a method, which allows provision of a larger variety of human milk oligosaccharides than prior art methods, preferably in a cost efficient manner, and preferably on an industrial scale.
Further, the provision of oligosaccharides and mixtures of oligosaccharides having between 4 and 12 saccharide units, such as between 6 and 10 saccharide units, in a stereoselective fashion and in a cost effective manner suitable to large scale production of oligosaccharides is desirable.